OpenAI Privacy Filter
About
OpenAI Privacy Filter is an Apache 2.0 open-weight bidirectional token-classification model for detecting and redacting personally identifiable information in text. OpenAI reports 1.5B total parameters, 50M active parameters, 128k-token context, and availability on Hugging Face and GitHub; this is a new specialist model release rather than a ChatGPT feature or endpoint alias.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution
Specifications
FamilyOpenAI Privacy Filter
Released2026-04-22
Parameters1.5B
Context128k
Architectureencoder_token_classifier
Specializationprivacy
LicenseApache 2.0
Trainingpretraining
Created by
Cutting-edge research and development.
San Francisco, California, United States
Founded 2015
Website